百度硅谷AI实验室：深度语音2：英语和普通话端到端语音识别.pdf

Deep Speech 2: End-to-End Speech Recognition in English and Mandarin Baidu Research – Silicon Valley AI Lab Dario Amodei, Rishita Anubhai, Eric Battenberg, Carl Case, Jared Casper, Bryan Catanzaro, 5 Jingdong Chen, Mike Chrzanowski, Adam Coates, Greg Diamos, Erich Elsen, Jesse Engel, 1 Linxi Fan, Christopher Fougner, Tony Han, Awni Hannun, Billy Jun, Patrick LeGresley, 0 Libby Lin, Sharan Narang, Andrew Ng, Sherjil Ozair, Ryan Prenger, Jonathan Raiman, 2 Sanjeev Satheesh, David Seetapun, Shubho Sengupta, Yi Wang, Zhiqian Wang, Chong Wang, c Bo Xiao, Dani Yogatama, Jun Zhan, Zhenyao Zhu e D Abstract 8 We show that an end-to-end deep learning approach can be used to recognize ] either English or Mandarin Chinese speech—two vastly different languages. Be- L cause it replaces entire pipelines of hand-engineered components with neural net- C works, end-to-end learning allows us to handle a diverse variety of speech includ- s. ing noisy environments, accents and different languages. Key to our approach is c our application of HPC techniques, resulting in a 7x speedup over our previous [ system [26]. Because of this efficiency, experiments that previously took weeks now run in days. This enables us to iterate more quickly to identify superior ar- 1 chitectures and algorithm